Extend filtered aggregation optimizer to support only masked partial aggregation cases

[feilong-liu]

Description

The current optimizer MergePartialAggregationsWithFilter only supports case where there are both aggregations with and without filter on the same filter. For example,
select sum(totalprice), sum(totalprice) filter (where orderpriority='2-HIGH') from orders group by orderdate
will work, but
select sum(totalprice) filter (where orderstatus='F'), sum(totalprice) filter (where orderpriority='2-HIGH') from orders group by orderdate
will not work.
In this change, I extend it to work for the second case too.

Motivation and Context

Extend existing optimizer to a broader range of queries.

Impact

Extend existing optimizer to a broader range of queries.

Test Plan

Add unit tests

Contributor checklist

• Please make sure your submission complies with our contributing guide, in particular code style and commit standards.
• PR description addresses the issue accurately and concisely. If the change is non-trivial, a GitHub Issue is referenced.
• Documented new properties (with its default value), SQL syntax, functions, or other functionality.
• If release notes are required, they follow the release notes guidelines.
• Adequate tests were added if applicable.
• CI passed.

Release Notes

Please follow release notes guidelines and fill in the release notes below.

== RELEASE NOTES ==

General Changes
* Extend  MergePartialAggregationsWithFilter to work for queries where all aggregations have mask

[jaystarshot]

I think the default case when one mask, one normal is always beneficial but this new case might not be beneficial for high cardinality cases so we should try to make the first one on by default later

[kaikalur]

I think the default case when one mask, one normal is always beneficial but this new case might not be beneficial for high cardinality cases so we should try to make the first one on by default later

But I hope adding the disjunction mitigates that? We will run some benchmarks and see.

[yhwang]

Hi @feilong-liu , I am working on the release notes for this PR. Can you revise the PR note and use more user-understandable wording?